A clinical deep learning framework for continually learning from cardiac signals across diseases, time, modalities, and institutions
- Resource Type
- article
- Authors
- Dani Kiyasseh; Tingting Zhu; David Clifton
- Source
- Nature Communications, Vol 12, Iss 1, Pp 1-11 (2021)
- Subject
- Science
- Language
- English
- ISSN
- 2041-1723
Deep learning algorithms trained on data streamed temporally from different clinical sites and from a multitude of physiological sensors are generally affected by a degradation in performance. To mitigate this, the authors propose a continual learning strategy that employs a replay buffer.